All of this got me thinking about the intersection of advertising and online content. I’ve been meaning to write something about this for some time now and Brown’s final question seemed like the perfect point to jump off from:

Wasn’t the web supposed to be better than this? Wasn’t there a promise that we could generate money and meaning, not merely the former? I’m still looking; but when I find it I will part with every penny I have.

I do believe that for some sites the answer has been found. It’s called the Deck. The Deck is an excellent example of an advertising platform that makes sense on the web. It succeeds in spite of (or perhaps because of?) its unobtrusiveness. I’ve found over a repeated period of time that I click on more ads it runs than on any other competing advertising platform, including Google’s.

For those not familiar with the Deck, a quick introduction. The Deck is an advertising network run by the design firm Coudal Partners. At the time of writing, it consists of a network of 33 selected web sites and services whose audience is primarily creative, web and design professionals (or those with pretensions to be so). Every one of those 33 sites and services shows only one advertisement per page and each advertisement consists of a 120 px x 90 px image with 180 characters of accompanying text. That’s it. The sites that are part of the Deck agree not to run any other advertisements so that Deck ad is the only ad on their page. At the time of writing, a month of advertising across the Deck will set you back a cool US$6900.

There are two reasons I believe the Deck works. The first is that they’ve found the balance between advertising and content. As Brown notes in her piece, as advertising becomes more disruptive we become better at tuning it out (either by avoiding the content that the ads are attached to or by selectively blocking out the ads using technological tools). But the Deck showcases an alternative way to get our attention: don’t be noisy, be exclusive.

The second reason the Deck works is that it’s targeted. Now when people think about targeted advertising they usually think about the kind of algorithmic targeting that has made Google a squillion dollars. But there is a different kind of targeting that for want of a better term I’ll call targeting by intelligent design. Targeting by intelligent design may be even more successful than algorithmic targeting.

Targeting by intelligent design has two aspects. The first is to have a sufficiently narrow audience. The audience of people that view the ads on the Deck are likely to be creative, web or design professionals because the sites that form the Deck are aimed specifically at that small segment. They’re almost all blogs run by a single blogger and almost all have a small range of stories they’ll cover³.

The second element of targeting by intelligent design has to do with the advertisers themselves. Advertisements on the Deck are only accepted if the Deck’s owners have either used or paid for that service or product. While at first glance this might seem counterproductive I would contend (and my experience demonstrates) it produces better, more relevant ads that are more likely to appeal to the target audience. Moreover, it elevates the ads from shill to endorsement since there is implicit in their placement the knowledge that these ads, and these ads alone, have met the standard necessary to be displayed.

Targeting by intelligent design is by no means an idea exclusive to the Deck. The web comic Penny Arcade successfuly employs a variation of the same strategy. The Penny Arcade web site features only two advertisements on their news pages and one ad on their comic pages. Although the ads typically feature animation and video they are tastefully located and tightly integrated into the overall design of the page. Penny Arcade has similar rules to the Deck and only carries advertisements for products or services they’ve used or paid for.

Intelligent design won’t work for everyone of course. The New York Times is read by too many different people to be able to employ exactly the same tactics as the Deck or Penny Arcade but perhaps a combination of the two is possible.

Consider the case of the news magazine Slate. Slate is an online-only magazine very much in the mould (when it comes to the range of content) of news magazines like Time and Newsweek. Its main focus is on US politics but it runs feature stories on science, health, economics and the environment as well as lighter fare on topics ranging from the latest movies to what to do in the event that you’ve slept with your wife’s sister.

Unfortunately for Slate, it employs the worst examples of algorithmic and intelligent design targeting. The main banner ads are certainly selected based on a rough idea of the type of people that constitute the audience but it is a very rough idea. While the target audience for Slate is no doubt narrowish it’s still a fair bit wider than that of Penny Arcade. Two ads that I’ve seen feature prominently in recent months have been for Hyundai cars and for a National Geographic special. Now I have no interest in Hyundai cars. I am neither in the market for a car nor am I sure I could even buy a Hyundai in Japan if I even wanted to. I’m also not particularly interested in National Geographic specials. Both these ad campaigns were completely wasted on me. They’re classy ads, to be sure, and they most assuredly cost a lot of money to make. The problem is I’m never going to click on them.

The algorithmic advertising is even worse. When I visit at the moment I’m bombared with offers to murder my wrinkles, whiten my teeth and gain certification in Six Sigma. I have no interest in clicking on any of these⁴. They distract from the intelligent design advertisements, make the site look cheap and, from a reading point of view, add a degree of clutter that makes it annoying to read the content I actually came to the site to view.

Now why can’t Slate combine the two and feature fewer ads but make them better targeted? The fact of the matter is no matter how wonderful an ad, I’m likely to only click on one advertisement per page (if only because once I’ve clicked on it I’m going to be taken away from the page). Before we had the sophisticated kind of algorithms we now do this might have justified showing a bunch of ads and hoping the law of averages meant I’d click on one of them. But we’re in 2009 now. Surely we can do better and select only one ad, but make it be one I’m almost certain to want to click.

Maybe it’s all too complicated once you get past a certain size. Perhaps large content-producing monoliths like newspapers and magazines are doomed to go the way of the Dodo and this is what Brown was getting at when she lamented the web’s seeming inability to combine making money with making meaning. That may be the case and if it comes to pass I’ll be disappointed for sure. But hey, you big guys had a good run for a while. Maybe it’s the turn of the small fry now.

The campaign began with two ads featuring Jerry Seinfeld and Bill Gates. In the ads Gates and Seinfeld engaged in comic banter while talking about almost everything except Windows. See the first ad here and the second one here. A number of people in the tech blogosphere opined that they didn’t get the whole thing. Michael Arrington:

I’m starting to feel bad for Microsoft PR, who’ve been tasked with defending these Microsoft ads featuring Bill Gates and Jerry Seinfeld. I just didn’t connect with the first ad, which barely mentioned Microsoft and didn’t do much to tell me why I should like their products in a competitive market. The second ad, which aired tonight, was more of the same.

There were of course defenders of the campaign. Mathew Ingram:

I made a marketing expert friend of mine mad recently when she said that the marketing professionals she knew didn’t like the original ad — and thought Microsoft was getting taken to the cleaners by its ad agency, Crispin Porter + Bogusky — because it was a dumb idea, or at least not a smart one. I tried to make the point that I don’t think Microsoft cares whether she and her marketing colleagues think the campaign is “smart” or not. They aren’t the target market any more than Mike Arrington is. I think whoever put these together is really just trying to humanize a giant company, and that’s a tough assignment.

The debate continued with the defenders claiming the ads were enjoyable diversions and the detractors pointing out that the goal of advertising is to sell. Then, just when the whole thing was starting to look Mexican standoffish, Microsoft announced there would be no more Seinfeld ads. Way to go, Redmond! Flame war averted!

You see it was all part of the plan, Microsoft PR assured us. This was just Phase 1 and the goal of Phase 1 was to start a conversation (everyone in marketing, please give us back that word). With that accomplished it was time to move on to Phase 2. Those of us not born yesterday were a little sceptical. John Gruber:

And dropping these ads is a panicked reaction. Let’s not pretend it makes any sense that the Seinfeld spots were planned as a two-episode teaser all along. No one signs Jerry Seinfeld for $10 million in a much-heralded deal to make just two spots that only run for a grand total of two weeks. The most telling fact is that the firm that reached out to the media yesterday to explain that this sudden shift was supposedly the plan all along was not Crispin Porter, the advertising agency producing the campaign, but Waggener Edstrom, Microsoft’s PR firm. Advertising campaigns which are going according to plan do not need PR firms to assert such.

So what’s Phase 2? Well, you know those ads from Apple where Justin Long pretends to be a Mac and John Hodgman a PC? Microsoft does and they think they’re unfair to people that use Windows computers. In an attempt to rectify the stereotype that people who use PCs all look like John Hodgman they’ve spliced together a lot of people who don’t look like John Hodgman (and one guy who eerily does).

In the interests of full disclosure, I was firmly in the camp of people who thought the initial advertisements were a failure. To be clear, I enjoyed them. Bill Gates and Jerry Seinfeld have real chemistry and if they return I’ll be more than happy to watch them get up to further hijinks. But advertisements are about selling things and these ads sold nothing: neither the Windows operating system nor the Windows brand.

Unfortunately, Phase 2 isn’t much better. It doesn’t take quite as long to sell you nothing but at the end of the day all Microsoft does in the ad is communicate to me that lots of people use Windows.

Memo to Microsoft: I already know this.

I am aware that Windows is ginormously successful. Everyone is aware Windows is ginormously successful. But you’re not selling cigarettes in the 1950s. You need to do something a little more sophisticated than argue a billion people can’t be wrong. You need to argue you product is reliable/interoperable/safe/fun/something. You need to argue that I want to use your product. (You cannot simply add sharks.)

Only today at work I watched some colleagues discover to their dismay that the printer they had just purchased didn’t work with their shiny MacBook Pro. This is not an uncommon occurrence and Microsoft’s entire campaign could be based around experiences similar to this. It is not impossible to sell the idea that Windows gets things done. Hey, that could even be the slogan! Windows: Get Things Done. It’s a simple message that explains why you want to use Windows (so you can get on with your life) while making the veiled accusation that competing products (read: Macs) are all about style with little substance to back it all up.

Maybe that’s Phase 3.

The logic behind the success of advertising to people using a search engine is relatively straightforward. You go to Google for one reason: you’re looking for something. If it’s something someone else can provide to you then they’re interested in getting your attention and (and really this is the crucial part) you’re interested in clicking their link. But if you go to Facebook or MySpace how interested are you in what someone is pushing on you? Regardless of how-well-targeted it is?

Advertising in social networks reminds me of when a friend recommends you go and see a movie. How successful is that ever? I can count the number of times on one hand I think someone’s said I ‘must see’ a film and I’ve actually done it. Certainly, I can think of plenty of times I’ve all but begged my friends to see a film and found that seemingly the more I’m interested in the movie the less they are. This goes for TV shows, music, podcasts and just about any other content you can think of. So difficult is it, in fact, that I’ve developed a tactic for suggesting things without suggesting thing; having them ‘come up’ in conversation and gently steering the person into thinking that they’re coming up with the idea of going.

Because isn’t that what it’s really about? We want to believe we made the choice. We don’t want to think that we do whatever someone else tells us to. And as soon as we start feeling they are our resistance levels increase dramatically.

I want to be clear to contrast this from asking for help from our friends. Here the recommendations are almost always acted upon but that’s because we’re choosing to elicit them them. Recommendations aren’t being pushed in this instance, they’re being pulled. When people talk about the huge influence our friends have on us, I tend to think they’re talking about their influence when we’ve decided it’s time to make a decision and we start looking for advice. Not when we’re trying to watch something on TV.

Which brings us back to the success of social networking advertising. Because how do you monetise pulled recommendations? If we jump onto the Flixster Facebook application, for instance, to see what our friends think we’re interested in what our friends think, not on ads being served up to us because of what our friends think. I guess if someone serves us up some cinema locations it’d be relevant, but that seems a much, much, much smaller potential pool of advertisers. And a correspondingly much, much, much smaller pool of money.